Machine learning electrocardiography model to differentiate takotsubo syndrome from myocardial infarction

Aug 5, 2026

Neural network models using an ECG approach are developed to distinguish takotsubo syndrome (TTS) from myocardial infarction (MI), including both ST-elevation and non-ST-elevation MI, as well as patients with suspected MI. A UNet-architecture network is trained and validated on 507 confirmed TTS cases and 14,978 controls from the Swedish coronary angiography and angioplasty register, with cross-validation and comparison against cardiologists’ previously proposed ECG criteria. Results show strong discrimination with high sensitivity and negative predictive value, outperforming conventional criteria, while further refinement and external validation are required to improve positive predictive value for screening and ruling out TTS.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages10

Reading Time25 min

AccessFree to download

Methods and results

Study design and data sources

Model performance and comparison

Conclusion

Aims

Introduction

European Heart Journal-Digital Health (2025) 6, 929–938 [https://doi.org/10.1093/ehjdh/ztaf073](https://doi.org/10.1093/ehjdh/ztaf073) ORIGINAL ARTICLE Machine learning electrocardiography model to differentiate takotsubo syndrome from myocardial infarction Felicia H. K. Hakansson 1,*, Erik Bodin2, Vincent Dutordoir2, Axel Gemvik1, Thomas Olsson1, Isabelle Nilsson3, Mikael Andersson Franko 1, Jonas Spaak 4 , Christina Ekenbäck 4, Loghman Henareh 5, Carl Henrik Ek 2, and Per Tornvall 1 1Department of Clinical Science and Education Södersjukhuset, Karolinska Institutet, Stockholm, Sweden; 2Department of Computer Science and Technology, University of Cambridge, Cambridge, UK; 3Department of Cardiology, Capio St. Görans Sjukhus, Stockholm, Sweden; 4Department of Clinical Sciences, Danderyds Hospital, Division of Cardiovascular Medicine, Karolinska Institutet, Stockholm, Sweden; and 5Department of Medicine, Huddinge, Karolinska Institutet, Stockholm, Sweden Received 25 February 2025; revised 12 April 2025; accepted 3 May 2025; online publish-ahead-of-print 23 June 2025 Methods and results Cross-sectional study in Stockholm. A neural network with UNet architecture was trained and validated on 507 TTS cases and 14 978 controls with suspected and verified MI, identified from the Swedish coronary angiography and angioplasty register. Cross-validation was performed. The models were compared with cardiologists using previously proposed ECG criteria. Receiver operating characteristics (ROC) area under the curve (AUC) for discriminating TTS from patients with STelevation and non-ST-elevation MI ROC AUC 0.88 (cross-validation: 0.85–0.92) and 0.86 (cross-validation: 0.82–0.91), respectively. ROC AUC for discriminating TTS from verified MI [non-ST-elevation MI (NSTEMI) and ST-elevation MI (STEMI)] was 0.87 (cross-validation: 0.83–0.91) with sensitivity (0.75) and specificity (0.83) with low positive predictive value (PPV) and high negative predictive value (NPV) . Results for suspected MI was ROC AUC 0.85 (cross validation: 0.81–0.91) with sensitivity (0.75) and specificity (0.79) with low PPV (0.11) and high NPV (0.99) . The committee of two cardiologists using a combination of ECG criteria achieved an ROC AUC of 0.71. Conclusion Machine learning models could discriminate TTS from MI (NSTEMI and STEMI) and suspected MI with high sensitivity and NPV, outperforming cardiologists using conventional criteria. The models require further refinement to increase PPV, precision-recall and external validation, but it holds promise for TTS screening aiding the clinician in ruling out TTS. Aims Machine learning (ML) algorithms applied to the electrocardiography (ECG) have been successful in several cardiac diagno ses, however, rarely been used for the diagnostics oftakotsubo syndrome (TTS). Our aim was to develop ML-based ECGmodels to differentiate TTS from patients with myocardial infarction (MI) . * Corresponding author. Tel: +46767726076, Email: [felicia.hakansson@ki.se](felicia.hakansson@ki.se) © The Author(s) 2025. Published by Oxford University Press on behalf of the European Society of Cardiology. This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial License ([https://creativecommons.org/licenses/by-nc/4.0/](https://creativecommons.org/licenses/by-nc/4.0/)), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact [reprints@oup.com](reprints@oup.com) for reprints and translation rights for reprints. All other permissions can be obtained through our RightsLink service via the Permissions link on the article page on our site—for further information please contact [journals.permissions@oup.com](journals.permissions@oup.com). Graphical Abstract Keywords Takotsubo syndrome • Myocardial infarction • Electrocardiography • Machine learning • Artificial

FAQs

What is the document’s main goal?

To develop machine learning–based ECG models that differentiate takotsubo syndrome from myocardial infarction and from patients with suspected MI.

How were the models trained and validated?

A neural network with UNet architecture was trained and validated using 507 TTS cases and 14,978 controls identified from the Swedish coronary angiography and angioplasty register, using cross-validation.

How did the machine learning models perform compared with cardiologists?

The models achieved higher ROC AUC values for TTS vs MI and suspected MI than cardiologists using conventional ECG criteria; cardiologist committee performance reported an ROC AUC of 0.71.

Aug 5, 2026

Uploaded by Seraphina